Angiotensin Inhibition Potentiates the Renal Responses to Neutral Endopeptidase Inhibition in Dogs with Congestive Heart Failure Kenneth B. Margulies, Mark A. Perrella, Linda J. McKinley, and John C. Burnett, Jr.
Cardiorenal Research Laboratory, Departments of Internal Medicine and Physiology, Mayo Clinic and Foundation, Rochester, Minnesota 55905
Abstract The renal natriuretic actions of endogenous atrial natriuretic factor are enhanced by neutral endopeptidase inhibition (NEPI). Recognizing that activation ofthe renin-angiotensin-aldosterone system in congestive heart failure (CHF) antagonizes the renal actions of atrial natriuretic factor, we hypothesized that angiotensin II antagonism with converting enzyme inhibition would potentiate the renal actions of NEP-I in CHF. To test this hypothesis, the renal responses to a specific NEP-I (SQ 28,603) were assessed in dogs with eight days of experimental CHF produced by rapid ventricular pacing. The renal natriuretic responses to NEP-I in experimental CHF were significant. In the same model of CHF, chronic angiotensin antagonism with converting enzyme inhibition potentiated both renal hemodynamic and excretory responses to NEP-I. The potentiated renal hemodynamic response included significant increases in glomerular filtration rate and filtration fraction. In the CHF group with angiotensin antagonism, an intrarenal infusion of low-dose angiotensin abolished the potentiated renal responses to NEP-I, supporting the concept that intrarenal angiotensin antagonism, rather than improved systemic hemodynamics or potentiation of other peptide systems, mediated the enhanced renal responses to NEP-I in the presence of converting enzyme inhibition. (J. Clin. Invest. 1991. 88:1636-1642.) Key words: converting enzyme inhibition * sodium excretion c renal hemodynamics * atrial natriuretic factor * rapid ventricular pacing
Introduction Atrial natriuretic factor (ANF)1 is a peptide hormone of cardiac origin with multiple biologic effects including natriuresis, diuresis, vasodilation, and inhibition of renin and aldosterone secretion. Among pathophysiologic states associated with eleThis study was presented in part at the American Society of Nephrology's 23rd annual meeting, Washington, DC, 2-5 December 1990, and was published in abstract form in 1990. (JASN [J. Am. Soc. Nephrol.]. 1:420.) Address correspondence to Kenneth B. Margulies, M.D., Cardiorenal Research Laboratory, Mayo Clinic and Foundation, 200 First Street SW, Rochester, MN 55905. Receivedfor publication 9 January 1991 and in revisedform S June 1991. 1. Abbreviations used in this paper: AII, angiotensin II; CEI, converting enzyme inhibition; CHF, congestive heart failure; Dist FRNa, distal fractional reabsorption of sodium; FF, filtration fraction; GFR, glomerular filtration rate; NEP-I, neutral endopeptidase inhibition; P1 ANF, plasma atrial natriuretic factor, Prox FRNa, proximal fractional sodium reabsorption; RAAS, renin-angiotensin-aldosterone system; UNaV, absolute sodium excretion. The Journal of Clinical Investigation, Inc. Volume 88, November 1991, 1636-1642 1636
vated circulating ANF, some of the highest levels are seen in humans and animals with congestive heart failure (CHF) (1). However, several studies (2-7) have demonstrated a blunted renal response to exogenous ANF in CHF. Multiple mechanisms have been implicated in this blunted renal response to ANF in CHF including: changes in ANF receptor density (810), enhanced degradation and clearance of ANF (11, 12), altered postreceptor signal transduction (13), activation of counterregulatory neurohumoral systems (14, 15), and decreased renal perfusion pressure (5). These mechanisms are not mutually exclusive. Accumulating evidence suggests that activation of the renin-angiotensin-aldosterone system (RAAS) in part contributes to sodium retention in CHF and renal resistance to ANF in this sodium avid state. In fact, previous studies have demonstrated antagonism between the RAAS and ANF with respect to both renal hemodynamics and tubular sodium handling (16, 17). In general, ANF promotes sodium excretion and vasorelaxation while angiotensin II (All) promotes sodium retention and vasoconstriction. Moreover, in a recent study by Raya et al. (18) All inhibition with a converting enzyme inhibitor enhanced hemodynamic responsiveness to ANF in rats with CHF, and a preliminary report indicates that All inhibition restored the natriuretic response to infused ANF in dogs with high-output CHF (15). In contrast, one recent study failed to demonstrate an association between RAAS activation and ANF resistance (19). Thus, most available evidence suggests that activation of the RAAS antagonizes the renal actions of ANF in CHF. ANF is degraded in vitro and in vivo by neutral endopeptidase (12, 20-22). Recent studies from our laboratory have demonstrated that neutral endopeptidase inhibition (NEP-I) potentiates the renal action ofboth physiologic and pathophysiologic levels of ANF in normal anesthetized dogs (22) and dogs with experimental CHF (12). While these results suggest a potential therapeutic role of NEP-I, the effects of NEP-I, like ANF itself, may be attenuated by activation of the RAAS in the setting of severe CHF. Therefore the current studies were designed to test the hypothesis that chronic inhibition of All generation in dogs with experimental CHF will potentiate the renal effects of NEP-I. To test this hypothesis, the renal responses to a specific NEP-I (SQ 28,603) were assessed in dogs with eight days of experimental CHF. All inhibition was achieved with five days of low dose oral converting enzyme inhibition (CEI) with captopril. To further define the role of All in modulating the renal action of NEP-I, exogenous All was administered intrarenally into one kidney of captopril treated dogs receiving acute NEP-I.
Methods Surgical preparation. Experiments were performed on two groups of six male mongrel dogs weighing between 15 and 24 kg and maintained
K. B. Margulies, M. A. Perella, L. J. McKinley, and J. C. Burnett, Jr.
on a fixed sodium intake of 100 mEq/d. All studies conformed to the guiding principles of the American Physiological Society. Experimental CHF was produced by ventricular pacing at 240 beats per min for 8 d before acute interventions. This experimental model has been shown consistently to produce a state of low output cardiac failure with the constellation of cardiac, renal, and neurohormonal abnormalities characteristic of chronic primary myocardial dysfunction (23, 24). In particular, this model is characterized by marked increases in circulating ANF and activation of the RAAS in association with avid sodium retention. Programmable pacemakers (model 8426; Medtronic Inc., Minneapolis, MN) were implanted 6 d before the start of pacing. Under pentobarbital anesthesia and via a left thoracotomy and pericardiectomy, the heart was exposed and a screw-in epicardial pacemaker lead was implanted into the right ventricular apex. The pacemaker lead was connected to a pulse generator implanted subcutaneously on the chest. The pericardium was sutured closed, and the parietal pleura and skin were closed in layers. The dogs were then allowed to recover over a 5-d period, the first 3 d of which they received prophylactic antibiotic treatment with clindamycin and CombioticO (Pfizer, Inc., New York, NY). 5 d after pacemaker implantation, the pacemakers were programmed to 240 beats per min and pacing continued for the remainder of the experiment. In group II, inhibition of All generation with captopril 12.5 mg three times daily (Bristol-Myers-Squibb, Inc., Princeton, NJ) commenced 3 d after the initiation of pacing. All dogs were given lithium carbonate 300 mg orally and fasted 16-20 h before the acute experiment. Dogs were permitted to drink water ad lib. until the time of the experiments. On the ninth day of pacing, dogs were anesthetized using fentanyl (0.005-0.01 mg/kg) and sodium pentobarbital (5-10 mg/kg intravenously). Supplemental 30-mg doses of pentobarbital were given as needed during the experiment. The animals were intubated and mechanically ventilated (Harvard respirator; Harvard Apparatus, Millis, MA) with supplemental oxygen at 4 liters/min. A femoral artery was cannulated for monitoring of arterial pressure and blood sampling. Femoral veins were cannulated for infusion of inulin and injection of the NEP-I and supplemental anesthetic. A balloon-tipped, flow-directed pulmonary artery catheter (model 93A- 3 1-7F; American Edwards Laboratories, AHS del Caribe, Inc., Anasco, Puerto Rico) was placed via the right internal jugular vein for measurement of cardiac filling pressures and determination of cardiac output. A flank incision was performed to gain exposure to the left kidney in all groups. The ureter was cannulated for urine collection and a noncannulating electromagnetic flow probe (Carolina Medical Electronics, Inc., King, NC) was placed around the renal artery for online measurement of renal blood flow. In group II, bilateral flank incisions were performed and both kidneys instrumented as described above. In addition, in all kidneys studied, curved 23-gauge needles, attached to polyvinyl tubing, were inserted into each renal artery distal to the flow probe and were kept patent by an infusion of 0.9% saline at 0.5 ml/min. Experimental protocol. A priming dose of inulin was given, and a constant infusion of I ml/min was begun in an effort to achieve a steady-state plasma level of 40-60 mg/dl. Following a 60-min equilibration period, a sample of arterial blood was drawn for baseline hormonal analysis. In group II an intrarenal infusion of All (1.5 ng/kg per min, Sigma Chemical Co., St. Louis, MO) was initiated into the right renal artery while the contralateral kidney and the instrumented kidney in
group I received the saline vehicle at the same rate (0.5 ml/min). The dose of All was selected to produce pathophysiologic levels of All in the supplemented kidney without affecting systemic hemodynamics or renal function in the contralateral kidney (16). After a 15-min lead in period, 30-min "pre-NEP-I" renal clearances were performed for all kidneys. During this and subsequent clearance periods, blood was drawn for hormonal analyses and determination ofinulin, sodium, and lithium concentrations. Cardiac output, mean arterial pressure, right and left atrial pressure, renal blood flow, and urine flow were also measured during each clearance period. Additional urine samples were collected on ice for measurement of urinary ANF and urinary cGMP. At the conclusion of the pre-NEP-I clearance period, in all groups
the NEP-I (SQ 28,603, Bristol-Myers-Squibb, Inc.) was given intravenously over 2 min at a dose of 30 mg/kg dissolved in 20 ml of 0.83% sodium bicarbonate solution. SQ 28,603 (N-[2-(mercaptomethyl)-1oxo-3-phenylpropyl]-f3-alanine) is a potent competitive inhibitor of neutral endopeptidase 24.11 with a Ki value of 1 nM. SQ 28,603 is selective in that it demonstrates only weak activity against other renal brush border peptidases such as angiotensin converting enzyme (I50 = 32 gM) and aminopeptidase M in which there was no inhibition at concentrations up to 1 IAM (Dalaney, N. G., unpublished data). Two additional 30-min clearance periods were performed beginning 15 min after the NEP-I was given. In all experiments, urinary fluid losses were replaced with intravenous normal saline every 15 min. Analysis. Cardiac output was determined by thermodilution (Cardiac Output model 9510-A computer, American Edwards Laboratories, Irvine, CA), measured in quadruplicate and averaged during each clearance period. Pulmonary capillary wedge pressure was used to determine left atrial pressure and right atrial pressure was measured directly. Systemic vascular resistance was calculated by dividing cardiac output by the difference between mean arterial and right atrial pressures. All voided urine was collected on ice, measured using a graduated cylinder, and aliquoted for measurement of sodium, inulin, lithium, ANF, and cGMP concentrations. Urine samples for ANF analysis were stored at -20'C until assay. Urine samples for cGMP determination were heated to > 90'C to eliminate phosphodiesterase activity and stored at -20'C until analysis. Blood for sodium, creatinine, and lithium determination was collected in heparinized tubes, placed on ice, and centrifuged at 2,500 rpm at 4VC. After centrifugation, plasma was separated and refrigerated until analysis. Glomerular filtration rate (GFR) was determined by inulin clearance. Plasma and urine inulin concentrations were determined by the anthrone method (25). To calculate renal filtration fraction, GFR was divided by renal plasma flow. Plasma and urinary sodium concentrations were measured using ion-selective electrodes (Beckman Instruments, Brea, CA), and plasma and urine lithium levels were determined by flame-emission spectrophotometry (model 357; Instrumentation Laboratory, Inc., Lexington, MA). Proximal fractional sodium reabsorption (Prox FRNa) was calculated using the formula: Prox FRNa (%) = [1 - (Li clearance/inulin clearance)] x 100. Distal fractional reabsorption of sodium (Dist FRNa) was calculated using the formula: Dist FRNa (%) = [(Li clearance-Na clearance)/(Li clearance)] X 100. Blood for hormone analysis was placed in EDTA tubes, immediately placed on ice, and centrifuged at 2,500 rpm at 4°C. Plasma was separated and stored at -20°C until assay. Extracted arterial plasma levels of ANF and urinary ANF were measured by radioimmunoassay to a-hANF as previously described (1). Plasma renin activity was determined by radioimmunoassay using the method of Haber et al. (26). Plasma aldosterone concentration was measured by specific radioimmunoassay (27). Plasma samples for cGMP were extracted with ethanol. Plasma and urinary cGMP were measured by radioimmunoassay using the method of Steiner (28). Fractional excretion of ANF was calculated by dividing the urinary clearance of ANF by the urinary clearance of inulin. Net renal generation of cGMP, a measure of nephrogenous cGMP production, was determined using the formula: Net renal generation of cGMP = (urinary cGMP X urine flow rate) - (plasma cGMP X GFR). All data are presented as mean±standard error. Within each group, comparisons with baseline were made using analysis of variance for repeated measures followed by Dunnett's t test when appropriate. Comparisons between groups were analyzed by one-way analysis ofvariance followed by Fisher's least significant difference test when appropriate. Comparisons of absolute changes after NEP-I between kidneys in group II were analyzed by paired Student's t test. Statistical significance was defined as P < 0.05.
Results At baseline, before initiation of intrarenal infusions, plasma
renin activity was moderately but significantly increased in the
Angiotensin Antagonism Potentiates Responses to Endopeptidase Inhibition
1637
captopril-treated group (n = 6) compared with untreated controls (n = 6) as illustrated in Fig. 1. The systemic hemodynamic and hormonal parameters for the two experimental groups before and after NEP-I administration are presented in Table I. In the pre-NEP-I period, mean plasma ANF levels were lower with inhibition ofAII generation in group II, yet plasma cGMP levels did not differ between the two groups. Renal hemodynamic and excretory parameters for the two experimental groups before and after NEP-I administration are presented in Table II. Before NEP-I administration, there were no significant differences in renal hemodynamic or excretory function between the CHF groups with and without inhibition of AII generation. In CHF with NEP-I alone (group I), the only significant cardiac hemodynamic response to NEP-I was a decrease in right atrial pressure observed at one hour post-NEP-I. In addition, a significant increase in plasma ANF was observed after NEP-I administration (524±77 to 680±104 pg/ml), which returned towards baseline levels 1 h post-NEP-I. Plasma cGMP levels paralleled this response. Sustained decreases in tubular sodium reabsorption were observed after NEP-I administration in group I, as indicated by increases in fractional excretion of sodium and decreases in distal fractional sodium reabsorption. However, there were no significant renal hemodynamic responses to NEP-I. In group II with CHF and inhibition of AII generation, there were no initial systemic hemodynamic responses to NEPI administration, but a small decrease in cardiac output and increased systemic vascular resistance were observed 1 h postNEP-I. In addition, a significant increase in plasma ANF was observed after NEP-I in group II (346±63 to 447±71 pg/ml), and plasma ANF levels tended to remain increased at 1 h postNEP-I administration. Plasma cGMP levels increased initially after NEP-I then returned to baseline levels. In contrast to group I, significant renal hemodynamic changes were observed in the CHF group with CEI after NEP-I administration. Specifically, a 58.7% increase in GFR, a 9.9% decrease in renal blood flow, and a 91.7% increase in filtration fraction were observed in group II, and these changes tended to be sustained. Increases in urine flow, absolute sodium excretion, and fractional excre-
12-
F
p
Cardiorenal Research Laboratory, Departments of Internal Medicine and Physiology, Mayo Clinic and Foundation, Rochester, Minnesota 55905
Abstract The renal natriuretic actions of endogenous atrial natriuretic factor are enhanced by neutral endopeptidase inhibition (NEPI). Recognizing that activation ofthe renin-angiotensin-aldosterone system in congestive heart failure (CHF) antagonizes the renal actions of atrial natriuretic factor, we hypothesized that angiotensin II antagonism with converting enzyme inhibition would potentiate the renal actions of NEP-I in CHF. To test this hypothesis, the renal responses to a specific NEP-I (SQ 28,603) were assessed in dogs with eight days of experimental CHF produced by rapid ventricular pacing. The renal natriuretic responses to NEP-I in experimental CHF were significant. In the same model of CHF, chronic angiotensin antagonism with converting enzyme inhibition potentiated both renal hemodynamic and excretory responses to NEP-I. The potentiated renal hemodynamic response included significant increases in glomerular filtration rate and filtration fraction. In the CHF group with angiotensin antagonism, an intrarenal infusion of low-dose angiotensin abolished the potentiated renal responses to NEP-I, supporting the concept that intrarenal angiotensin antagonism, rather than improved systemic hemodynamics or potentiation of other peptide systems, mediated the enhanced renal responses to NEP-I in the presence of converting enzyme inhibition. (J. Clin. Invest. 1991. 88:1636-1642.) Key words: converting enzyme inhibition * sodium excretion c renal hemodynamics * atrial natriuretic factor * rapid ventricular pacing
Introduction Atrial natriuretic factor (ANF)1 is a peptide hormone of cardiac origin with multiple biologic effects including natriuresis, diuresis, vasodilation, and inhibition of renin and aldosterone secretion. Among pathophysiologic states associated with eleThis study was presented in part at the American Society of Nephrology's 23rd annual meeting, Washington, DC, 2-5 December 1990, and was published in abstract form in 1990. (JASN [J. Am. Soc. Nephrol.]. 1:420.) Address correspondence to Kenneth B. Margulies, M.D., Cardiorenal Research Laboratory, Mayo Clinic and Foundation, 200 First Street SW, Rochester, MN 55905. Receivedfor publication 9 January 1991 and in revisedform S June 1991. 1. Abbreviations used in this paper: AII, angiotensin II; CEI, converting enzyme inhibition; CHF, congestive heart failure; Dist FRNa, distal fractional reabsorption of sodium; FF, filtration fraction; GFR, glomerular filtration rate; NEP-I, neutral endopeptidase inhibition; P1 ANF, plasma atrial natriuretic factor, Prox FRNa, proximal fractional sodium reabsorption; RAAS, renin-angiotensin-aldosterone system; UNaV, absolute sodium excretion. The Journal of Clinical Investigation, Inc. Volume 88, November 1991, 1636-1642 1636
vated circulating ANF, some of the highest levels are seen in humans and animals with congestive heart failure (CHF) (1). However, several studies (2-7) have demonstrated a blunted renal response to exogenous ANF in CHF. Multiple mechanisms have been implicated in this blunted renal response to ANF in CHF including: changes in ANF receptor density (810), enhanced degradation and clearance of ANF (11, 12), altered postreceptor signal transduction (13), activation of counterregulatory neurohumoral systems (14, 15), and decreased renal perfusion pressure (5). These mechanisms are not mutually exclusive. Accumulating evidence suggests that activation of the renin-angiotensin-aldosterone system (RAAS) in part contributes to sodium retention in CHF and renal resistance to ANF in this sodium avid state. In fact, previous studies have demonstrated antagonism between the RAAS and ANF with respect to both renal hemodynamics and tubular sodium handling (16, 17). In general, ANF promotes sodium excretion and vasorelaxation while angiotensin II (All) promotes sodium retention and vasoconstriction. Moreover, in a recent study by Raya et al. (18) All inhibition with a converting enzyme inhibitor enhanced hemodynamic responsiveness to ANF in rats with CHF, and a preliminary report indicates that All inhibition restored the natriuretic response to infused ANF in dogs with high-output CHF (15). In contrast, one recent study failed to demonstrate an association between RAAS activation and ANF resistance (19). Thus, most available evidence suggests that activation of the RAAS antagonizes the renal actions of ANF in CHF. ANF is degraded in vitro and in vivo by neutral endopeptidase (12, 20-22). Recent studies from our laboratory have demonstrated that neutral endopeptidase inhibition (NEP-I) potentiates the renal action ofboth physiologic and pathophysiologic levels of ANF in normal anesthetized dogs (22) and dogs with experimental CHF (12). While these results suggest a potential therapeutic role of NEP-I, the effects of NEP-I, like ANF itself, may be attenuated by activation of the RAAS in the setting of severe CHF. Therefore the current studies were designed to test the hypothesis that chronic inhibition of All generation in dogs with experimental CHF will potentiate the renal effects of NEP-I. To test this hypothesis, the renal responses to a specific NEP-I (SQ 28,603) were assessed in dogs with eight days of experimental CHF. All inhibition was achieved with five days of low dose oral converting enzyme inhibition (CEI) with captopril. To further define the role of All in modulating the renal action of NEP-I, exogenous All was administered intrarenally into one kidney of captopril treated dogs receiving acute NEP-I.
Methods Surgical preparation. Experiments were performed on two groups of six male mongrel dogs weighing between 15 and 24 kg and maintained
K. B. Margulies, M. A. Perella, L. J. McKinley, and J. C. Burnett, Jr.
on a fixed sodium intake of 100 mEq/d. All studies conformed to the guiding principles of the American Physiological Society. Experimental CHF was produced by ventricular pacing at 240 beats per min for 8 d before acute interventions. This experimental model has been shown consistently to produce a state of low output cardiac failure with the constellation of cardiac, renal, and neurohormonal abnormalities characteristic of chronic primary myocardial dysfunction (23, 24). In particular, this model is characterized by marked increases in circulating ANF and activation of the RAAS in association with avid sodium retention. Programmable pacemakers (model 8426; Medtronic Inc., Minneapolis, MN) were implanted 6 d before the start of pacing. Under pentobarbital anesthesia and via a left thoracotomy and pericardiectomy, the heart was exposed and a screw-in epicardial pacemaker lead was implanted into the right ventricular apex. The pacemaker lead was connected to a pulse generator implanted subcutaneously on the chest. The pericardium was sutured closed, and the parietal pleura and skin were closed in layers. The dogs were then allowed to recover over a 5-d period, the first 3 d of which they received prophylactic antibiotic treatment with clindamycin and CombioticO (Pfizer, Inc., New York, NY). 5 d after pacemaker implantation, the pacemakers were programmed to 240 beats per min and pacing continued for the remainder of the experiment. In group II, inhibition of All generation with captopril 12.5 mg three times daily (Bristol-Myers-Squibb, Inc., Princeton, NJ) commenced 3 d after the initiation of pacing. All dogs were given lithium carbonate 300 mg orally and fasted 16-20 h before the acute experiment. Dogs were permitted to drink water ad lib. until the time of the experiments. On the ninth day of pacing, dogs were anesthetized using fentanyl (0.005-0.01 mg/kg) and sodium pentobarbital (5-10 mg/kg intravenously). Supplemental 30-mg doses of pentobarbital were given as needed during the experiment. The animals were intubated and mechanically ventilated (Harvard respirator; Harvard Apparatus, Millis, MA) with supplemental oxygen at 4 liters/min. A femoral artery was cannulated for monitoring of arterial pressure and blood sampling. Femoral veins were cannulated for infusion of inulin and injection of the NEP-I and supplemental anesthetic. A balloon-tipped, flow-directed pulmonary artery catheter (model 93A- 3 1-7F; American Edwards Laboratories, AHS del Caribe, Inc., Anasco, Puerto Rico) was placed via the right internal jugular vein for measurement of cardiac filling pressures and determination of cardiac output. A flank incision was performed to gain exposure to the left kidney in all groups. The ureter was cannulated for urine collection and a noncannulating electromagnetic flow probe (Carolina Medical Electronics, Inc., King, NC) was placed around the renal artery for online measurement of renal blood flow. In group II, bilateral flank incisions were performed and both kidneys instrumented as described above. In addition, in all kidneys studied, curved 23-gauge needles, attached to polyvinyl tubing, were inserted into each renal artery distal to the flow probe and were kept patent by an infusion of 0.9% saline at 0.5 ml/min. Experimental protocol. A priming dose of inulin was given, and a constant infusion of I ml/min was begun in an effort to achieve a steady-state plasma level of 40-60 mg/dl. Following a 60-min equilibration period, a sample of arterial blood was drawn for baseline hormonal analysis. In group II an intrarenal infusion of All (1.5 ng/kg per min, Sigma Chemical Co., St. Louis, MO) was initiated into the right renal artery while the contralateral kidney and the instrumented kidney in
group I received the saline vehicle at the same rate (0.5 ml/min). The dose of All was selected to produce pathophysiologic levels of All in the supplemented kidney without affecting systemic hemodynamics or renal function in the contralateral kidney (16). After a 15-min lead in period, 30-min "pre-NEP-I" renal clearances were performed for all kidneys. During this and subsequent clearance periods, blood was drawn for hormonal analyses and determination ofinulin, sodium, and lithium concentrations. Cardiac output, mean arterial pressure, right and left atrial pressure, renal blood flow, and urine flow were also measured during each clearance period. Additional urine samples were collected on ice for measurement of urinary ANF and urinary cGMP. At the conclusion of the pre-NEP-I clearance period, in all groups
the NEP-I (SQ 28,603, Bristol-Myers-Squibb, Inc.) was given intravenously over 2 min at a dose of 30 mg/kg dissolved in 20 ml of 0.83% sodium bicarbonate solution. SQ 28,603 (N-[2-(mercaptomethyl)-1oxo-3-phenylpropyl]-f3-alanine) is a potent competitive inhibitor of neutral endopeptidase 24.11 with a Ki value of 1 nM. SQ 28,603 is selective in that it demonstrates only weak activity against other renal brush border peptidases such as angiotensin converting enzyme (I50 = 32 gM) and aminopeptidase M in which there was no inhibition at concentrations up to 1 IAM (Dalaney, N. G., unpublished data). Two additional 30-min clearance periods were performed beginning 15 min after the NEP-I was given. In all experiments, urinary fluid losses were replaced with intravenous normal saline every 15 min. Analysis. Cardiac output was determined by thermodilution (Cardiac Output model 9510-A computer, American Edwards Laboratories, Irvine, CA), measured in quadruplicate and averaged during each clearance period. Pulmonary capillary wedge pressure was used to determine left atrial pressure and right atrial pressure was measured directly. Systemic vascular resistance was calculated by dividing cardiac output by the difference between mean arterial and right atrial pressures. All voided urine was collected on ice, measured using a graduated cylinder, and aliquoted for measurement of sodium, inulin, lithium, ANF, and cGMP concentrations. Urine samples for ANF analysis were stored at -20'C until assay. Urine samples for cGMP determination were heated to > 90'C to eliminate phosphodiesterase activity and stored at -20'C until analysis. Blood for sodium, creatinine, and lithium determination was collected in heparinized tubes, placed on ice, and centrifuged at 2,500 rpm at 4VC. After centrifugation, plasma was separated and refrigerated until analysis. Glomerular filtration rate (GFR) was determined by inulin clearance. Plasma and urine inulin concentrations were determined by the anthrone method (25). To calculate renal filtration fraction, GFR was divided by renal plasma flow. Plasma and urinary sodium concentrations were measured using ion-selective electrodes (Beckman Instruments, Brea, CA), and plasma and urine lithium levels were determined by flame-emission spectrophotometry (model 357; Instrumentation Laboratory, Inc., Lexington, MA). Proximal fractional sodium reabsorption (Prox FRNa) was calculated using the formula: Prox FRNa (%) = [1 - (Li clearance/inulin clearance)] x 100. Distal fractional reabsorption of sodium (Dist FRNa) was calculated using the formula: Dist FRNa (%) = [(Li clearance-Na clearance)/(Li clearance)] X 100. Blood for hormone analysis was placed in EDTA tubes, immediately placed on ice, and centrifuged at 2,500 rpm at 4°C. Plasma was separated and stored at -20°C until assay. Extracted arterial plasma levels of ANF and urinary ANF were measured by radioimmunoassay to a-hANF as previously described (1). Plasma renin activity was determined by radioimmunoassay using the method of Haber et al. (26). Plasma aldosterone concentration was measured by specific radioimmunoassay (27). Plasma samples for cGMP were extracted with ethanol. Plasma and urinary cGMP were measured by radioimmunoassay using the method of Steiner (28). Fractional excretion of ANF was calculated by dividing the urinary clearance of ANF by the urinary clearance of inulin. Net renal generation of cGMP, a measure of nephrogenous cGMP production, was determined using the formula: Net renal generation of cGMP = (urinary cGMP X urine flow rate) - (plasma cGMP X GFR). All data are presented as mean±standard error. Within each group, comparisons with baseline were made using analysis of variance for repeated measures followed by Dunnett's t test when appropriate. Comparisons between groups were analyzed by one-way analysis ofvariance followed by Fisher's least significant difference test when appropriate. Comparisons of absolute changes after NEP-I between kidneys in group II were analyzed by paired Student's t test. Statistical significance was defined as P < 0.05.
Results At baseline, before initiation of intrarenal infusions, plasma
renin activity was moderately but significantly increased in the
Angiotensin Antagonism Potentiates Responses to Endopeptidase Inhibition
1637
captopril-treated group (n = 6) compared with untreated controls (n = 6) as illustrated in Fig. 1. The systemic hemodynamic and hormonal parameters for the two experimental groups before and after NEP-I administration are presented in Table I. In the pre-NEP-I period, mean plasma ANF levels were lower with inhibition ofAII generation in group II, yet plasma cGMP levels did not differ between the two groups. Renal hemodynamic and excretory parameters for the two experimental groups before and after NEP-I administration are presented in Table II. Before NEP-I administration, there were no significant differences in renal hemodynamic or excretory function between the CHF groups with and without inhibition of AII generation. In CHF with NEP-I alone (group I), the only significant cardiac hemodynamic response to NEP-I was a decrease in right atrial pressure observed at one hour post-NEP-I. In addition, a significant increase in plasma ANF was observed after NEP-I administration (524±77 to 680±104 pg/ml), which returned towards baseline levels 1 h post-NEP-I. Plasma cGMP levels paralleled this response. Sustained decreases in tubular sodium reabsorption were observed after NEP-I administration in group I, as indicated by increases in fractional excretion of sodium and decreases in distal fractional sodium reabsorption. However, there were no significant renal hemodynamic responses to NEP-I. In group II with CHF and inhibition of AII generation, there were no initial systemic hemodynamic responses to NEPI administration, but a small decrease in cardiac output and increased systemic vascular resistance were observed 1 h postNEP-I. In addition, a significant increase in plasma ANF was observed after NEP-I in group II (346±63 to 447±71 pg/ml), and plasma ANF levels tended to remain increased at 1 h postNEP-I administration. Plasma cGMP levels increased initially after NEP-I then returned to baseline levels. In contrast to group I, significant renal hemodynamic changes were observed in the CHF group with CEI after NEP-I administration. Specifically, a 58.7% increase in GFR, a 9.9% decrease in renal blood flow, and a 91.7% increase in filtration fraction were observed in group II, and these changes tended to be sustained. Increases in urine flow, absolute sodium excretion, and fractional excre-
12-
F
p
